ERIC Number: ED190607
Record Type: RIE
Publication Date: 1980-Apr
Reference Count: 0
A Comparison of Various MRA Methods Applied to Longitudinal Evaluation Studies in Vocational Education.
Kapes, Jerome T.; And Others
Three models of multiple regression analysis (MRA): single equation, commonality analysis, and path analysis, were applied to longitudinal data from the Pennsylvania Vocational Development Study. Variables influencing weekly income of vocational education students one year after high school graduation were examined: grade point averages (grades 7-9, and 10-12), socioeconomic status, and initial income. All models concluded that earnings of vocational graduates did not differ significantly from nonvocational graduates; furthermore, initial income was the best and only predictor of income one year after graduation. The path model also suggested that vocational education affects achievement in grades 10-12, which in turn affects initial income. While the three models yielded the same overall coefficient of determination, commonality and path models provided more information about the unique and shared variance of predictor variables. Path models, in particular, were recommended because they also reflected direct and indirect influence of all other antecedent or suppressor variables. (CP)
Descriptors: Grade Point Average, High School Graduates, High Schools, Income, Intelligence, Least Squares Statistics, Longitudinal Studies, Multiple Regression Analysis, Path Analysis, Postsecondary Education, Predictor Variables, Program Evaluation, Socioeconomic Status, Suppressor Variables, Vocational Education
Publication Type: Speeches/Meeting Papers; Reports - Research
Education Level: N/A
Authoring Institution: N/A
Identifiers: Commonality Analysis
Note: Paper presented at the Annual Meeting of the American Educational Research Association (64th, Boston, MA, April 7-11, 1980).